Title CLASSIFICATION AND BIOLOGY OF JAPANESE INSECTIVORA (MAMMALIA) : Ⅱ. Biological Aspects Author(s) Abe, Hisashi Citation Journal of the Faculty of Agriculture, Hokkaido University, 55(4), 429-458 Issue Date 1968-05 Doc URL http://hdl.handle.net/2115/12830 Type bulletin (article) File Information 55(4)_p429-458.pdf Instructions for use Hokkaido University Collection of Scholarly and Academic Papers : HUSCAP CLASSIFICATION AND BIOLOGY OF JAPANESE INSECTIVORA (MAMMALIA) II. Biological Aspects Hisashi ABE (Natural History Museum, Faculty of Agriculture Hokkaido Uni\'ersity, Sapporo) Recei\'ed December ri, 1967 As a continuation of the first paper of this work (1) dealing with the variation and classification of the Japanese insectivores, the present paper com­ prises the ecological accounts of their habitat, food habits, population structure and reproductive activity. The material and method are accordingly the same as in the former paper. (1) Sorcx lIJzguicuiatlis DOHSO:\' and Sorex C(lecuticllS saez'lIs TIIOMAS Habitat. Sorc.x llllguicuiatlis (big-clawed shrew) is common in grassy meadows and shrublands ranging from lowlands to low mountain regions; it is less abundant in woods of mature forest type with little undergrowth. Sorex caecutiens saevus (ezo shinto shrew) is most abundant in the shrub­ lands and the woods of mountain regions (Table 1). In the woods it is not so abundant as in the shrublands, but still distinctly predominates over the big-clawed shrew. The ezo shinto shrew is scarcely seen in grassy meadows where the big-clawed shrevy is dominant. TABLE 1. Differences of occurrence between the two speCIes of SoreL in various vegetations Type of vegetation S. UJl,!j'uiOl{atlfs S. caecut;ellS S(lC'Z'llS Forests e 8Ui ,~ HU " ;' Bushe~ (Shrubs>Grasses) :2!).~ 73.8 Bushes (Shrubs<Grasses) I 75.S 21.2 Grasses 91i,O 4.0 In areas where the forest has been removed, either by logging or by fire, the land is revegetated by shrubs, young trees, sasa bamboos, and herbs. In such areas shrews are usually very abundant; the big-clawed shrew is rather I Jour. Facul. Agr. Hokkaido Gni\'., Sapporo, \'01. 55. Pt. 4. 1%81 430 H. ABE abundant in the bushes having a rich understory of grasses, while the ezo shinto shrew is more common in bushes in which young trees and sasa bamboos are dominant. In moderate brushes suitable for both the species, it was never seen that the two species lived together in numbers. If one species was abundant, the other was present only in small numbers or was absent. This suggests the existence of some interspecific intolerance between the two species. The habitat of the shrews is usually rich in ground litter, providing shelter and hunting areas. This is also the same as regards certain wild mice. The ezo shinto shrew associates occasionally with forest-dwelling mice, such as ApodclllllS argcJltcus TDI:\lI:\CK, and the big-clawed shrew with Clethrionomys rufocanus bcdfordiae (TIlOl\L\S), which also inhabits mainly shrub lands and grassy meadows. Although both species of shrews are inhabitants of subterranean runways, the ezo shinto shrew is not a digger. On the other hand, the big-clawed shrew appears to be a semidigger. This may be demonstrated by their food habits and morphological characters; i. e., the former has a relatively long tail, large hind feet, longer ears, small hands and claws, and feeds much on spiders, whereas the latter has a relatively short tail, small hind feet, short ears, very large hands with long claws, and feeds much on earthworms. From these facts, the big-clawed shrew appears to be more modified morphologically for a fossorial existence than does the ezo shinto shrew. Abundant occurrence of this semifossorial shrew, S. lmguiculatlls, in the grass fields and shrublands of Hokkaido may be caused partly by the absence of fossorial mammals such as moles and shrew-moles which are more robust and prefer a habitat similar to that of this shrew. After the thawing of snow in spring, often are found many small holes on the surface of ground in the big-clawed shrew's range. The holes, measur­ ing about 2.5 cm in diameter and 3 to 5 cm in depth, seem to have been dug by the big-clawed shrew for hunting food, and around them are small piles of earth containing the remains of insects and snails, probably eaten by the shrews. In the range of the ezo shinto shrew, no such holes are found; if any holes are made, they probably be obscured because of the heavily littered habitat. Food habits. Stomachs and alimentary tracts of 66 big-clawed shrews and 117 ezo shinto shrews collected in all seasons were examined. In the big­ clawed shrews, insects make up more than one-half of the food, and next in abundance are earthworms, centipedes, root stocks, seeds, molluscs, and arach­ nids (Table 2). The insects identified comprise two orders: Coleoptera and Lepidoptera. The large bulk and frequency of earthworms and the small CLASSIFICATION A::,\:D BIOLOGY OF JAPANESE INSECTIVORA 431 TABLE 2. Stomach contents of 66 individuals of SorcI unguiculatus Percentage by Percentage by bulk occurrence I. Insecta 59.2 a. Coleoptera 2.5 7.6 b. Lepidoptera 2.3 3.0 c. Undetermined 54.4 68.2 II. Arachnida 1.4 9.1 Ill. Chilopoda Geophilus 5.7 16.7 IV. Gastropoda :2.2 10.6 V. Chaetopoda Lumbricomorpha :'::2.5 40.9 VI. Undetermined (animals) 2.9 3.0 VII. Hair 0.0 1.5 VIII. Plants Seeds 2.8 3.0 Root stocks 2.9 :).0 (,') Dead grasses ).- 9.1 IX. Sand 0.2 1.5 percentage of spiders are distinctly different from what have been found in the ezo shinto shrew. In the latter, almost all of the stomachs contain insects, which compose by bulk the largest share of food. Next in order of abundance by bulk are arachnids, centipedes, seeds, molluscs, earthworms, crustaceans, and millipedes (Table 3). This composition of stomach contents is very charac­ teristic in that there is a large percentage of arachnids contrasted ",ith a small percentage of earthworms, compared with the food of the big-cla\\-ed shrew. This difference in food habits is significant, for the respective tendencies are evident even in individuals of both species obtained in the same habitat during the same season. This has probably resulted from the different habits of the two species; namely, it indicates more fossorial activity of the big-clawed shrew. The proportion of vegetable matter by bulk is very small in the two species, and these data well conform to many other reports on the related but different species of Sorcx (3, 4). It has been said that shrews prey on wild mice, but the author strongly 432 II. ABE TABLE 3. Stomach contents of 117 individuals of Sorex caecutiells saevus Percentage by Percentage by bulk occurrence I. Insecta Li5.:) a. Coleoptera O.:l 1.7 b. Diptera 0.1; 1.7 c. Lepidoptera l.~ 1.7 I""') d. Undetermined ).) ...... 88.0 II. .-\rachnida lCJ.9 35.9 III. Chilopoda Geophilus 1O.S 25.fJ IV. Gastropoda o.s 1.7 V. Crustacea Isopoda O.:l 0.9 VI. Chaetopoda Lumbricomorpha 0.5 2.ii VII. l: ndeterminecl (animals) 0:2 0.9 \'II I. Hair 0.1 0.9 IX. Plants Seeds 2.0 3.4 Dead grasses O.~ 0.9 doubts whether the shrews concerned prey on healthy wild mice under natural conditions. To elucidate this question, a small red-backed vole, Clethrio71omys rutilus mikado, was placed in a cage with seven big-cla\ved shrews, and they \vere kept alive for two days. During this period nothing appeared to happen to any of these animals. Although captive shrews devour the carcasses of mice, they probably do not kill healthy mice. The absence of remains of any other mammal in the stomachs also suggests that these shrews do not prey naturally on small mammals. The average weight of daily food was 18 grams for a captive young big­ clawed shrew weighing 11 grams. This IS more than one and half times of its body weight. Population structure. In early spring. v"hen the breeding actIVIty does not yet begin. populations of these shrews are composed of members of a single age-class, namely the over-wintered adults born in the previous year. Its (l t"' > TABLE 4. Seasonal variation in the population structure in C/l C/l .... the two species of Sorex '1l () Month Jan. Feb. Mar. Apr. May Jun. Jui. Aug. Sep. Oct. Nov. Dec. ~ ~ > a. Young 0 0 0 0 11 19 15 ':J7 , 16 62 2 0 Z ... tj ~ ..:l b. Subadult 9 12 10 32 2 2 0 1 1 0 0 0 to ::! (5 l-l .;: c. Adult 0 0 0 0 12 14 5 18 4 6 1 0 ot"' ~ ~ Q r;s § Per cent of 100 100 100 100 56.0 45.7 25.0 16.4 23.8 8.8 ? 0 ><: subadult & adult o '1l .... a . Young 0 0 0 0 0 1 9 53 41 58 17 2 .§'" ~ ..... ::! b. Subadult 1 0 0 0 0 0 0 ~ 19 4 11 2 3 tIl " C/l '"~ tIl c. Adult 0 0 0 0 1 4 1 7 1 3 1 0 Ii ~ ~ e Z Per cent of 100 100 83.3 10.0 11.7 2.4 4.9 5.6 0 C/l ~ 51 subadult & adult 100 100 100 tIl Q o::: :;d > ~ "" H. ABE simple structure changes into two sharply different age classes in late spring when new young are born, and this condition exists till November.